Re: [RFC] new substitute path when loading feature

[Aleksandar Ristovski <aristovski at qnx dot com>]

Daniel Jacobowitz wrote:
> Does this fix the same problem you're working on?  Does anyone see a
> problem with the compromises I made here?  Eli, I'd appreciate your
> opinion.

I was considering something like this (except I changed the IS_ABSOLUTE_PATH macro to a be defined as a function call) but then didn't like the idea... there was also opposition to that approach when I brought it up on the mailing list.

But yes, this can be a solution too, although I expect that unix purists will complain about the possibility of having directory names with colon on POSIX file systems (GDB_IS_ABSOLUTE_PATH could return wrong result in such cases). The same stands for backslashes.

To sum it up: in our case it (probably - haven't tested) does fix our common-scenario, but not sure if it's a 'proper' solution - probably not; maybe if it was wrapped in an option as was suggested in my first attempt at 'universal' IS_ABSOLUTE_PATH?